PulseAugur
LIVE 01:42:35
research · [8 sources] ·
0
research

New research benchmarks and methods advance Graph Neural Network evaluation and design

Several recent arXiv papers explore advancements and challenges in Graph Neural Networks (GNNs). Research includes methods for verifying GNN ownership and detecting copycat models, as well as developing unified benchmarks for evaluating knowledge graph construction and GNN performance. Other work focuses on standardized evaluation protocols for adversarial GNNs, surveys on graph rewiring techniques to mitigate over-squashing and over-smoothing, adaptive node feature selection, and a new algebraic framework called Grothendieck Graph Neural Networks that moves beyond the traditional neighborhood-based aggregation. AI

Summary written by gemini-2.5-flash-lite from 8 sources. How we write summaries →

IMPACT These papers collectively advance the understanding and capabilities of GNNs, potentially leading to more robust, interpretable, and powerful graph-based AI systems.

RANK_REASON Multiple arXiv papers published on various aspects of Graph Neural Networks, including verification, benchmarking, adversarial robustness, and architectural improvements.

Read on arXiv cs.LG →

COVERAGE [8]

  1. arXiv cs.LG TIER_1 · Rahul Nandakumar, Deepayan Chakrabarti ·

    COPYCOP: Ownership Verification for Graph Neural Networks

    arXiv:2605.05360v1 Announce Type: new Abstract: Given two GNNs that output node embeddings, how can we determine if they were trained independently? An adversary could have trained one GNN specifically to mimic the other GNN's embeddings. To obscure this relationship between the …

  2. arXiv cs.LG TIER_1 · Tran Gia Bao Ngo, Zulfikar Alom, Federico Errica, Murat Kantarcioglu, Cuneyt Gurcan Akcora ·

    Adversarial Graph Neural Network Benchmarks: Towards Practical and Fair Evaluation

    arXiv:2605.05534v1 Announce Type: new Abstract: Adversarial learning and the robustness of Graph Neural Networks (GNNs) are topics of widespread interest in the machine learning community, as documented by the number of adversarial attacks and defenses designed for these purposes…

  3. arXiv cs.LG TIER_1 · Othmane Kabal, Mounira Harzallah, Fabrice Guillet, Hideaki Takeda, Ryutaro Ichise ·

    A Unified Benchmark for Evaluating Knowledge Graph Construction Methods and Graph Neural Networks

    arXiv:2605.05476v1 Announce Type: new Abstract: Knowledge graphs automatically constructed from text are increasingly used in real-world applications. However, their inherent noise, fragmentation, and semantic inconsistencies significantly affect the performance of Graph Neural N…

  4. arXiv cs.LG TIER_1 · Hugo Attali, Nathalie Pernelle, Davide Buscaldi, Fragkiskos D. Malliaros ·

    Graph Rewiring in GNNs to Mitigate Over-Squashing and Over-Smoothing: A Survey

    arXiv:2605.00951v1 Announce Type: new Abstract: Graph Neural Networks are powerful models for learning from graph-structured data, yet their effectiveness is often limited by two critical challenges: over-squashing, where information from distant nodes is excessively compressed, …

  5. arXiv cs.LG TIER_1 · Ali Azizpour, Madeline Navarro, Santiago Segarra ·

    Adaptive Node Feature Selection For Graph Neural Networks

    arXiv:2510.03096v2 Announce Type: replace Abstract: We propose an adaptive node feature selection approach for graph neural networks (GNNs) that identifies and removes unnecessary features during training. The ability to measure how features contribute to model output is key for …

  6. arXiv cs.LG TIER_1 · Hugo Attali, Davide Buscaldi, Nathalie Pernelle, Fragkiskos D. Malliaros ·

    Graph Rewiring in GNNs to Mitigate Over-Squashing and Over-Smoothing: A Survey

    arXiv:2411.17429v2 Announce Type: replace Abstract: Graph Neural Networks are powerful models for learning from graph-structured data, yet their effectiveness is often limited by two critical challenges: over-squashing, where information from distant nodes is excessively compress…

  7. arXiv cs.AI TIER_1 · Kieran Maguire, Srinandan Dasmahapatra ·

    Mini-Batch Class Composition Bias in Link Prediction

    arXiv:2604.25978v1 Announce Type: cross Abstract: Prior work on node classification has shown that Graph Neural Networks (GNNs) can learn representations that transfer across graphs, when underlying graph properties are shared. For a fixed graph, one would then expect GNNs traine…

  8. arXiv cs.LG TIER_1 · Amirreza Shiralinasab Langari, Leila Yeganeh, Kim Khoa Nguyen ·

    Grothendieck Graph Neural Networks Framework: An Algebraic Platform for Crafting Topology-Aware GNNs

    arXiv:2412.08835v2 Announce Type: replace Abstract: Graph Neural Networks (GNNs) are almost universally built on a single primitive: the neighborhood. Regardless of architectural variations, message passing ultimately aggregates over neighborhoods, which intrinsically limits expr…